[OSM-talk] Historical Data in OSM database

[Richard Palmer]

Dear list,

  Many thanks for all the responses. I'll try to summarise them as I 
  understand them:

  o 	At the moment storing the information in the main database would/may
	cause problems for some OSM tools, so would be best to keep it
	separate for this reason alone.
  
  o	Running one "historical OSM" should be enough rather than lots of
	them. Funding permitting we might be able to run this (on a trial
	basis) if nobody else wanted too?.

  o	Work would be needed on the various OSM tools to support time
	dimension, and also on visualising the data.

  o	A historical OSM and the current OSM could be brought together again
	by using object IDs (and we could do regular syncs of the databases
	to add new buildings from the main database to the historical one).

  o	There is a gap that will emerge from changes to the main database
	if dating information isn't added to those objects.

  A few people pointed out the various difficuties due to changes in building
  shape/use over time and gaps in records and details. This is all true, but 
  the intention would not be to claim we have mapped Tudor England down to
  the last gibbet, but where we have some information on areas that can 
  be entered in the map and connected to information on the same area in later
  periods (and disputed, and updated, and annotated). 

  At the moment for many academic projects small maps are researched and
  drawn up to illustrate the history of a particular area, but the
  information cannot then be connected to anything else and often disappears
  when the project server is switched off.

regards,

Richard

-- 
Richard Palmer			| Centre for E-Research &
Systems Manager			| Centre for Computing in the Humanities
richard.d.palmer at kcl.ac.uk	| King's College London
Tel: 0207 848 1973